Skip to content

Data corruption for NCLOB fields #200

@papersaltserver

Description

@papersaltserver

Hello,
we're working on a CAP project and decided to use @sap/cdx-mtx library for multitenancy. Our project heavily uses CSV files to seed initial data.
It was found that cds-mtx writes CSV data with hdb library to HDI Container to NCLOB field, then it reads the same CSV data and hdb library returns this data corrupted. For some files (every time the same), but not all some data has repeating symbols, e.g. 100002 instead of 10002, Serbiaan instead of Serbian. I've tried to open TENANT_FILES table with Database Explorer and exactly in place of symbol duplication DE truncates the field. I've also tried to open the same table with DBeaver and data looks fine in the database itself.
How to debug this? What should we check?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions